Proceedings of the International Seminar on Mineral Processing Technology - 2006,

Chennai, India. pp. 3 - 10.

Beneficiation of Indian Heavy Mineral Sands — Some New

Possibilities Identified by Tata Steel
Ch. V.G.K. Murty, D. Rathod, S. Asokan and Amit Chatterjee.
Tata Steel Limited

Abstract

Titanium, the 9th most abundant element on the earth's crust, is available mostly (90%) in the
form of ilmenite (FeaTi02) and leucoxene (weathered ilmenite). These titanium-bearing
minerals occur either as placer minerals or some times; as rock deposits (e.g. in Canada and
Norway).
As a part of its vision to grow in the mineral and metal related industries, Tata Steel is looking
at business opportunities in the field of titania. The titania business will involve stage-wise
development of mining, separation of ilmenite as well as other co-minerals, beneficiation of
ilmenite to synthetic rutile, TiO, pigment production and ultimately the production of titanium
metal. The 'Teri' mineral sands (inland iron coated dunes) of Sattankulam in Tamil Nadu will
be the feed raw material.
Since normally in the heavy mineral sand industry, it is customary to pre-concentrate the
material to produce heavy mineral concentrate (HMC). While primary concentration is carried
out employing gravity concentration techniques to separate the heavy minerals (ilmenite,
rutile, zircon, leucoxene, sillimanite and monazite) from sand (quartz) based on their
significant differences in specific gravity, individual separation of minerals relies on
techniques such as electrostatic separation, magnetic separation and flotation. However,
difficult-to-treat type of sand deposits such the Teri deposits of Sattankulam, which are
characterised by a wide size range, higher proportion of slimes, and the presence of near
gravity materials like kyanite and corundum, necessitate an altogether different approach for
pre-concentration. The all-spiral-circuit that is normally used has to be supplemented by free
settling classifiers (for feed preparation) and hindered settling classifiers for the removal of
fine sillimanite).
While the use of electrostatic separators for separation of HMC into conducting and non-
conducting fractions is traditionally used at the initial.stage of dry milling in the magnetic
separation circuit, the recent trend is to focus on the advantages of using rare earth dry
magnetic separators in place of electrostatic separators for the separation of ilmenite,
especially in the case of ilmenite-rich deposits such as the one at Sattankulam.
This paper describes some new possibilities that have been identified by Tata Steel for the
development of appropriate flow sheets for the beneficiation of the Sattankulam heavy mineral
sands deposits.
INTRODUCTION
As a part of the Steel Company's vision to grow outside steel, Tata Steel is studying the possibility of
exploiting the extensive titania deposits in Tamil Nadu. It is planned that this venture, to be taken up
in phases would involve mining of the inland sand dunes followed by the separation of ilmenite and

3
 Beneficiation of Indian Heavy Mineral Sands — Some New Possibilities Identified by Tata Steel

other valuable minerals, up-gradation of the ilmenite to synthetic rutile, production of titanium dioxide
pigment and ultimately the production of titanium metal.
In its effort to utilise the Sattankulam deposits, Tata Steel has entered into a Memorandum of
Understanding (MoU) with the Tamil Nadu Government according to which, the Tamil Nadu
Government is providing the support including mineral rights and environmental clearance required to
begin work on this venture. To be -able to undertake the task of exploiting the titania deposits in
Sattankulam, Tata Steel has signed an agreement (in the month of September, 2003) with an
international consortium headed by Outokumpu to conduct a bankable feasibility study, including
geological assessment of the deposits over an area of 80 square kilometers granted to Tata Steel under
a Prospecting License. Besides Outokumpu, the consortium comprises PAH and L&T; Dasturco has
been appointed as the consultant for the project; and TZMI (TZ Minerals International) of Australia,
were selected to advise on the process to be selected for optimum utilisation of the deposit and
marketing of the product (s).
The entire feasibility study was split into two parts — Part A comprised the geological evaluation of
the resources, including drilling, resource evaluation, mining concepts and mine planning,
metallurgical test work, flow sheet development and an evaluation of the suitability of the ilmenite for
beneficiation, while Part B consisted of pilot plant tests for mineral separation, validation and
optimisation of flow sheets, environmental studies and optimisation of mine scheduling. Part A of the
feasibility study was completed in December 2004 and Part B in February 2006.
LOCATION
The heavy mineral resources to be utilised in Tata Steel's titania venture are located near Sattankulam
and Kuttam villages in the Tuticorin and Tirunelveli districts of Tamil Nadu. The Sattankulam
deposit is located at a distance of about 60 km SE of Tirunelveli town (Fig.1); the port town of
Tuticorin is located at a distance of about 80 km NE of Sattankulam thereby providing opportunities
for export of the finished product(s).

Fig. 1: Map of the Sattankulam Region in Tamil Nadu

4
 Proceedings of the International Seminar on Mineral Processing Technology

GEOLOGY OF THE DEPOSIT

The Sattankulam deposit is made up of very homogeneous grains, 85% of which is between 2mm and
63 micron in size and the remaining 15% is fines below 63 microns. The particles have a bright red
colour owing to an iron oxide coating on the surface. The thickness of the sand column varies between
0.8m and 15.7m with an average thickness of 6.9m. The contact zone between the bottom weathered
sandstone and the sand column is gradational( ')
Based on the slime content as well as the colour of the sand, two layers have been identified within the
sand column both of which have similar heavy mineral contents and mineral assemblage distribution,
but the distribution of fines is somewhat different. The upper sand layer having a thickness of upto 4m
consists of brick-red unconsolidated sand and generally has a low content of fines (11.5%). On the
other hand, the lower sand layer with thickness varying from 0 to 6 m is medium brown; is
unconsolidated and contains higher amount of fines (16.7%).(2) A typical geological section showing
the upper and lower sand columns along with the profile of the bottom sand stone formation is given
in Fig. 2.
MINERAL CHARACTERISATION
Mineral sands contain ilmenite, rutile, leucoxene, sillimanite, garnet, kyanite, monazite, zircon and
other co-minerals. The average total heavy mineral (THM) and slime (below 63 microns) contents
within the Sattankulam deposit are 10.27 % and 15.30 % respectively; the heavy minerals being
concentrated in the -0.25 to +0.075 mm size fraction. (3)
The typical heavy mineral assemblage in the THM is given below:
Minerals % of HM
llmenite 63.7 40M
Rutile 4.3 Upper Sand
30M
Zircon 4.3 .
Sillimanite 15.4 20M
Leucoxene 2.1 ItIM Lower and
Monazite 1.0
Garnet 1.3 00M
Magnetite 0.5
Fig. 2: Typical Geological Section of the Upper and Lower
Others 7.4 Sand Columns

Most of the ilmenite separates efficiently in the 3000 to 6000 gauss fraction, and altered ilmenite in
the 7000 to 10000 gauss fraction; the size of ilmenite falls between 75um and 212 pm, the majority
being in the size range of 150 to 106 pm. A preliminary mineralogical study indicated that the
ilmenite is present in the following forms:
■ Iron-rich ilmenite, where stoichiometrically excess iron in the form of exsolved hematite as well as
fine intergrowths with ilmenite mineral is present.
■ Ferro-magnetic (ferrian) ilmenite (easily attracted by a hand magnet with about 500 gauss) —
probably this ilmenite contains magnetite in ultra fine intergrowths.
• Stoichiometric ilmenite with a Ti: Fe ratio of around 1:1.16.
■ Titania-rich ilmenite with no exsolved hematite phase.
It'was also found that:
■ The MnO content of the ilmenite is low compared with ilmenites found elsewhere.
• Cr203 occurs in altered ilmenite but only in traces (7,000-10,000 gauss).
• V205 is evenly distributed in all the sizes and in all the magnetic fractions.

5
 Beneficiation of Indian Heavy Mineral Sands — Some New Possibilities Identified by Tata Steel

■
The A1203 and Si02 contents indicated the presence of either sillimanite or garnet in minor
amounts as a locked mineral or as separate particles in juxtaposition with ilmenite.
■
Rutile and zircon are relatively fine — over 75% of rutile and over 65% of zircon are below 150
micron in size.
PRIMARY CONCENTRATION PLANT (PCP) — WET CIRCUIT
The development of the flow sheet for the PCP was undertaken on the basis of achieving the best
product quality coupled with maximum possible recovery and the envisaged production levels of
ilmenite, rutile and zircon. Various processes of wet gravity concentration circuits were evaluated for
fulfilling these requirements. Traditionally, an all-spiral-circuit (Fig. 3) or a spiral-plus-table
combination is employed in primary concentration plants (wet circuit) to produce heavy mineral
concentrates (HMC).
Such equipment operate effectively as long as the ore body does not contain fine sillimanite, kyanite,
corundum, garnet, etc. Since the Sattankulam deposit contains about 16% sillimanite in fine form in
the total heavy minerals, it is essential to devise an effective method of removal of sillimanite at the
first opportunity; otherwise, it would make the mineral separation circuit quite complex. Therefore,
the use of hindered settling classifiers such as a Floatex separator was considered in the test work
conducted for the development of the flow sheet.(4)
The test work was designed to compare the recovery and grade of valuable heavy minerals (VHM) in
an all-spiral-circuit with a spirals-plus-Floatex circuit in which the spirals are used as the rougher and
scavenger and the Floatex density separator is used for cleaning as shown in Fig.4.

Mid Spiral

Fig. 3: Normal All Spiral Flow Sheet used for Fig. 4: Primary Concentration Plant Flow Sheet
Primary Concentration using Spirals-Plus-Floatex Separator
The results obtained in the study using both sets of equipment are summarised in Table 1.
The data in Table 1 shows that:
■
The Floatex separator recovered higher amounts of magnetic TiO2 and Zr02 — the underflow
contained as much as 99% of the Ti and Zr minerals. This underflow contained equal or higher
percentages of TiO2 or Zr02 compared with the spiral concentrate.
■
The Floatex separator underflow product contained only a small percentage of non-valuable heavy
minerals — ilmenite, zircon, rutile and leucoxene accounted for 96.6% of the HMC; only 1.4 % of
the underflow was non-valuable heavy minerals.
■
The Floatex separator overflow analysed 87.9% quartz and 12.1% of heavy minerals. The latter
contained 2.8% ilmenite and 8.7% sillimanite (corresponding to 71.5 % of the total sillimanite).

6
 Proceedings of the International Seminar on Mineral Processing Technology

Table 1: VHM Results Using an All-Spiral-Circuit Vis-A-Vis Spiral-Plus-Floatex Combination

Non-mag Non-
Mag TiO2 lye, Distribution Zr02
TiO2
Products Wt% (Ilmenite), of mag TiO2 mag dist
dist.,
(Rutile), Zr02 TiO2
% %
dist., %
Spirals
Conc T1 73.6 50.31 40.43 25.01 87.8 80.6 91.1
Mid T1 7.5 42.87 26.21 6.63 6.8 9.6 4.5
Tails T1 19.0 17.65 5.67 1.41 5.4 9.8 4.4
Conc T2 78.9 50.17 41.28 24.23 92.8 87.1 93.8
Mid T2 5.4 , 36.41 17.86 2.91 3.7 6.6 2.0
Tails T2 15.7 14.68 4.14 1.54 3.6 6.3 4.2
Conc T3 83.1 49.86 40.37 22.97 96.2 92.5 97.9
Mid T3 5.2 33.13 12.73 1.10 2.8 5.7 0.9
Tails T3 11.8 7.90 1.14 0.38 1.0 1.8 1.2
Floatex
U/F 83.6 50.29 42.36 26.94 98.9 98.9 99.3
0/F 16.4 3.76 0.79 0.35 1.1 1.0 0.7
A comparison of the recovery of VHM in the case of the all-spiral-circuit and the spirals-plus- Floatex
combination is presented in Table 2.
Based on the test work and computer simulations, it was evident that:
■
When the Floatex density separator is used in the final "cleaning" stage of concentration, it is more
effective than spirals.
■
The Floatex separator generates HMC that is higher in both grade and recovery percentages.
■
It also allows for single point control and can be readily adjusted unlike the adjustment of
numerous spiral splitters.
Keeping the superior performance of the spirals-plus-Floatex separator combination in view, the same
has been chosen for inclusion in the wet circuit. A distinct advantage of this combination is that
sillimanite will be concentrated into the Floatex overflow stream, thereby reducing the load on the
main dry plant.
ATTRITIONING AND SCRUBBING OF HEAVY IRON OXIDE COATINGS
One of the unique features of the Sattankulam deposit is that all the minerals have a coating of iron
oxide on them as shown in Fig. 5. The coating is either thick or thin - silicate minerals, especially
quartz form the predominant thick coatings over minerals like ilmenite, while the quartz coating on

Table 2: Comparison of Recovery of VHM in the Iron Coatings

Two Options
LII.Inanit6 prang without costing/I
Spirals- ar\-
All
Parameter plus-
spirals
Floatex
VHM grade, % 94 91
Recovery, %
I lmenite 94 90
Rutile 89 87
Fig. 5: Microphotograph Showing Iron Oxide
Zircon 94 94
Coating on HMC

7
 Reneficiation of Indian Heavy Mineral Sands — Some New Possibilities Identified by Tata Steel

the lighter particles is relatively thin. Such particles can be rejected by gravity separation without
requiring any treatment to remove the coating (leaving only the heavy mineral fractions to be treated).
To remove the heavy coating, HMC first has to be attritioned at high density using process water,
which removes most of the surface coating thereby rendering the HMC sufficiently clean to produce
high quality products in the dry mill. On the other hand, rutile and zircon production requires an
additional surface cleaning step using water, NaOH and H2SO4 as the scrubbing agents. Laboratory
and pilot scale test work conducted on attrition and scrubbing revealed the following:
■ There was little improvement when the scrubbing time was increased from 6 to 9 minutes.
■ Attrition scrubbing with water effectively liberated the individual minerals from the agglomerated
particles; it also removed some of the surface coating.
■ Scrubbing with NaOH (0.5 kg/tonne) and H2SO4 (10 kg/tonne) followed by
■ magnetic separation and electrostatic separation of the non-magnetic fraction revealed that the
scrubbing agent used to attrition the HMC affects the efficiency of the subsequent electrostatic
separation as shown in Fig. 6.
The cumulative TiO2 distribution for both acid (H2SO4) and caustic (NaOH) treatments was almost
identical for the conductor as well as the non-conductor products. Fig. 6 also shows that there was less
zircon in compartments 1-7 for the acid scrub and therefore, more zircon in the non-conductor
product.
■
For the recovery of Ti02, there was little difference between the two agents (see Table 3) — the
slightly higher TiO2 in the case of NaOH was within permissible sampling/ analytical errors.
Cumulative % Distribution

- H2SO4 ii02 Table 3: Test Data on Attrition and Scrubbing

—s— H2SO4 Zr02 Non
- NaOH TiO2 Conductor
conductor
onductor
- NaOH 7102
TiO2 ZrO2
Wt% Wt%
dist dist
2 3 4 5 6 7 NaOH 46.6 71.0 32.5 76.5
Compartment No H2SO4 45.4 69.6 35.6 84.9

Fig. 6: Cumulative Distribution of T102 and Zr02 in

Electrostatic Separation

■
For zircon, however, acid scrubbing gave better recovery results — 84.9% compared with 76.5%
for NaOH; the grade of non-conductors was the same at 24.9% Zr02 , in both cases.
MINERAL SEPARATION PLANT (MSP)
Ilmenite Circuit — Dry Magnetic Separation Vis-A-Vis Wet High Intensity Magnetic
Separation/Electrostatic Separation
While electrostatic separation is the first step in treating HMC after drying in India, normal flow
sheets in Eastern Australia employ wet high intensity magnetic separators (WHIMS) to produce
magnetic (ilmenite-rich) and non-magnetic concentrates. During the 1970s, Readings developed
electromagnetic versions of WHIMS that have been used in many operating mines along the East
coast of Australia. WHIMS has also been installed in the South African Sands Project and there is a
high probability that WHIMS will be employed in the Corridor project (when it comes up). On the
other hand, many plants in Western Australia and in the US do not use WHIMS — the HMC is dried

8
 Proceedings of the International Seminar on Mineral Processing Technology

and processed using rare earth drum (RED) magnets. Though a few operations continue to use dry
electromagnets, in the past ten years the use of dry electromagnets has been largely superseded by
vastly improved rare earth magnetic separators.
WHIMS is used when there is a preferential market for non-magnetic products such as rutile and
zircon over ilmenite. The various advantages of WHIMS are as follows:(5)
■
Unusable material can be discarded using WHIMS before any drying — this is of great advantage,
particularly with increasing fuel prices.
■
WHIMS is generally suitable for extracting primary ilmenite from HMC. Once ilmenite begins to
weather and loses its high magnetic susceptibility, WHIMS becomes less efficient in holding the
ilmenite in the magnetics stream.
■
In many MSP flow sheets, the zircon wet gravity circuit immediately follows the ilmenite removal step.
In such cases, the use of WHIMS obviates the necessity to dry the non-magnetics fraction twice.
The advantages of using a dry magnetic circuit include:
■
Such a circuit can be installed at a lower capital cost than a WHIMS circuit.
■
More controlled partitioning is possible because the separation medium is air, rather than water
and it is easier to move the separating particles apart in air.
■
In the laboratory under ideal conditions, separation of primary ilmenite from non-magnetics often
exhibits similar performance for both types of magnets. Industry experience, however, has shown
that the performance of REDs is easier to control and maintain than that of WHIMS, the
performance of the latter depending on the level of operator attention, water quality and
maintenance.
■
Dry magnetic separation circuits are simpler to operate and the maintenance cost is lower.
WHIMS, on the other hand, has many more wear parts and requires more complicated feed
pumping and distribution equipment.
Based on the above comparison, it has been decided that Tata Steel's project in Tamil Nadu would use
REDs. 100% drying of HMC followed by the use of REDs would provide the optimum solution for
the Sattankulam ilmenites.
ZIRCON CIRCUIT - HOT ACID LEACHING
The sand particles in the Sattankulam deposit have a heavy coating of iron oxide and the production of
premium (Fe203 < 0.1%) grade zircon would call for hot acid leaching, which has been found to be the
most effective for removing the coating.
The non conducting fraction from the ruffle circuit, which contains the crude zircon concentrate, will
be subjected to hot acid leaching followed by de-sliming at -63 microns (Fig. 7). The zircon
concentrate will be dried at 120°C before being further processed in an electrostatic separator unit to
remove the remaining TiO2 minerals, and a magnetic separator to remove the magnetic aluminum
silicates and monazite.
As shown in Table 4, hot acid leaching can help reduce Fe203 from 0.15 to 0.07%, thereby rendering
it a premium grade product.
CONCLUSION
In the heavy minerals industry it is vital to produce high quality saleable products to meet the ever
stringent demands of customers. Tata Steel's proposed titania venture in Tamil Nadu envisages
mining of raw sand in the Sattankulam deposit, for an initial annual production of 500,000 tonnes of
ilmenite, 34,000 tonnes of rutile and 22,000 tonnes of zircon. This would be followed in stages by the
production of 130,000 tonnes of synthetic rutile (if found feasible) and ultimately, 60,000 tonnes of
TiO, pigment.

9
 Beneficiation of Indian Heavy Mineral Sands Some New Possibilities Identified by Tata Steel

RUTILE/ZIRCON CIRCUIT

RED Non Mag

Spiral/ Floatex

Concentrate Tails
4. Table 4: Effect of Hot Acid Leaching on Fe203
Diving
Content of Concentrate
Oxide Before leaching After leaching
Zr02 64.4 66.6
Conductor Non Clnductor
TiO2 0.75 0.08
Crude Rutile Hot AddLeaching A1203 2.21 0.29
for further
processing Deslirning Fe203 0.15 0.07
Neutralization

Crude Mrcon
for further
processing

Fig. 7 Hot Acid Leaching in the Wet Zircon Circuit

Detailed geological, mineralogical and process option studies have been carried out to arrive at the
optimum flow sheet for the raw material feed. It has been found that the difficult to treat type of
deposits in the Teri deposits of Sattankulam are characterised by a wide size range of minerals, higher
proportion of slimes and the presence of near gravity products like kyanite and corundum, all of which
necessitate an altogether different approach if optimum exploitation has to be carried out. Dry mining
using a combination of front end loaders and conveyors/trucks has been chosen as the mining method.
A combination of spirals and Floatex separators has shown better recovery compared with an all
spirals circuit in pilot plant trials along with the advantage of sillimanite separation at a very early
stage. A combination of dry magnetic and high tension electrostatic separators is envisaged for
ilmenite production. While very clean rutile can be produced in a traditional rutile circuit employing
high tension separators, the crude zircon concentrate would require hot acid leaching (because of
heavy iron oxide coatings) to produce zircon of premium grade for ceramic applications.
ACKNOWLEDGEMENT
The authors would like to thank Dr.S.Mohan Rao of the R&D, Tata Steel and Ms. V. Hema Malini of
TBU, Tata Steel for their help and support in preparing this paper.
REFERENCES
[I] Rita A.M.Gardner, Dept. of geography, Kings College, London — Quaternary Coastal Sediments
& Stratigraphy, Southeast India, ISPQS-Man & Environment Vol.X (10), 1986, pp 51-72.
[2] Pincock Allen & Holt, USA, Report -Tata Integrated Titania Project Part A — Preliminary
Feasibility Study, Tirunelveli, Tamil Nadu, India, December'04 and Septermber'05 interim
reports.
[3] S Mohan Rao, R. Sripriya, K. Chandrakala, P. Panigrahi and Ch.V.G.K. Murty, Internal Report -
Characterisation and Beneficiation Studies of Mineral Sands from Tamil Nadu, March (2005).
[4] Ch.V.G.K.Murty, Dr. S.Asokan, D.Rathod, Development of Sattankulam Heavy Mineral Sands,
Tamilnadu, India by Tata Steel, Heavy Minerals 2005, Society for Mining, Metallurgy &
Exploration, 2005, pp 73-82.
[5] Steve Gilman, (2005), Dry Magnetic Separation Vis-à-vis Wet High Intensity Magnetic
Separation, Internal Communication, July 2005.

I0